• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.10
• /
• pp.3230-3255
• /
• 2022

Causality mining in NLP is a significant area of interest, which benefits in many daily life applications, including decision making, business risk management, question answering, future event prediction, scenario generation, and information retrieval. Mining those causalities was a challenging and open problem for the prior non-statistical and statistical techniques using web sources that required hand-crafted linguistics patterns for feature engineering, which were subject to domain knowledge and required much human effort. Those studies overlooked implicit, ambiguous, and heterogeneous causality and focused on explicit causality mining. In contrast to statistical and non-statistical approaches, we present Bidirectional Encoder Representations from Transformers (BERT) integrated with Multi-level Feature Networks (MFN) for causality recognition, called BERT+MFN for causality recognition in noisy and informal web datasets without human-designed features. In our model, MFN consists of a three-column knowledge-oriented network (TC-KN), bi-LSTM, and Relation Network (RN) that mine causality information at the segment level. BERT captures semantic features at the word level. We perform experiments on Alternative Lexicalization (AltLexes) datasets. The experimental outcomes show that our model outperforms baseline causality and text mining techniques.

• Journal of Biomedical Engineering Research
• /
• v.43 no.2
• /
• pp.109-115
• /
• 2022

Purpose: Medical records classification using vectorization techniques plays an important role in natural language processing. The purpose of this study was to investigate proper vectorization techniques for electronic medical records classification. Material and methods: 403 electronic medical documents were extracted retrospectively and classified using the cosine similarity calculated by Scikit-learn (Python module for machine learning) in Jupyter Notebook. Vectors for medical documents were produced by three different vectorization techniques (TF-IDF, latent sematic analysis and Word2Vec) and the classification precisions for three vectorization techniques were evaluated. The Kruskal-Wallis test was used to determine if there was a significant difference among three vectorization techniques. Results: 403 medical documents were relevant to 41 different diseases and the average number of documents per diagnosis was 9.83 (standard deviation=3.46). The classification precisions for three vectorization techniques were 0.78 (TF-IDF), 0.87 (LSA) and 0.79 (Word2Vec). There was a statistically significant difference among three vectorization techniques. Conclusions: The results suggest that removing irrelevant information (LSA) is more efficient vectorization technique than modifying weights of vectorization models (TF-IDF, Word2Vec) for medical documents classification.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.39 no.4
• /
• pp.67-71
• /
• 2016

The purpose of this study was to explore olfactory sensitivity of odors from injection molding processing. To do this, the experiment was carried out in an injection molding factory, and participants were exposed to the environment where odor-substances (Formaldehyde, Benzene, Toluene, Ethylbenzene, Xylene, Styrene) exist. In addition, we used the subjective scale using semantic adjectives as an olfactory sensitivity. As a result, the assessment structure of olfactory sensitivity was composed of eight factors (irritant, thermal, tense, unique, like-dislike, active, stable, masculine), and the main factor which was irritant characteristics explained 20% of the total olfactory sensibility. These results suggested that odors from injection molding processing would cause more negative emotional responses than the flavor which is mainly used in olfactory sensitivity. This study, as a basic study of the improvement in a factory environment for the efficiency of work, has limits in that it was conducted to the extent of identifying the olfactory sensitivity structure of those who were at a laboratory and who were exposed to the environment of odor substance induced in the injection molding processing. Therefore, for the method of removing malodorous substance, the effect of materials which can neutralize it, and the comparison of a direct performance in the environment where negative sensitivity structure exists, a series of studies which aim to improve the environment of injection molding factories, such as performance assessment in the environment of a factory and an office need to be conducted. It is expected that when these studies are put together, the improvement guidelines will be provided as a type that can maximize the effectiveness of work in the factory environment where injection molding processing is done.

• Journal of the Ergonomics Society of Korea
• /
• v.31 no.1
• /
• pp.49-61
• /
• 2012

Objective: The aim of this study is to offer a holism based Service Experience Analysis System(HSEAS) for service design. Background: Customer experience has been focused in a lot of service area. But it is difficult to understand the customer's needs and their experiences because it's so complex and uncertain. Thus it needs holistic approach that means it's difficult to explain general character merely through the understanding of small parts that composes an object and it must be analyzed within the overall context. Method: Accordingly, the thesis paper proposes the Service Experience Analysis System that satisfies the four following needs. (1) Need of solid Experience Framework in which the special quality of the service experience is considered, (2) need of support for the semantic cohesion between different kinds of data, (3) need of support for the management and search of vast data, and (4) need of building the knowledge base system for collaborative research. Results: HSEAS combines the short information in the customers' words and behaviors or situations and circumstances and provides a place of analysis where the context of the general experience can be read and allows concrete understanding of the actual state and factor of the problem as a Combined Data Analysis Tool. Conclusion: HSEAS becomes the center of information management, analysis and connection and it provides a free collaboration place where physical condition has no relations to as a knowledge base system based on network. Application: It is expected that length and width will be added to the analysis and assistance for effectively accumulating information will be provided in the area of diverse service.

• Journal of the Korean Society for information Management
• /
• v.29 no.4
• /
• pp.273-294
• /
• 2012

Images have been substantially searched and used due to not only the advanced internet and digital technologies but the characteristics of a younger generation. The purpose of this study aims to discuss the ways on expanding the access points to images by analyzing the needs of users in context of everyday life. In order to achieve the purpose of this study, 105 questions of image seeking in NAVER, which is one of social Q&A services in Korea, were analyzed. For the analysis, a two-dimensional framework with image uses and image attributes were utilized. The findings of this study demonstrate that considerable use purposes on data oriented pole, such as information processing, information dissemination and learning are identified. On the other hand, image attributes from the needs of image show that non-visual aspects including contextual attributes are recognized substantially in addition to the traditional semantic attributes.

• Journal of Korea Multimedia Society
• /
• v.7 no.10
• /
• pp.1452-1463
• /
• 2004

The requirement of automatic video summary is increasing as bi-directional broadcasting contents and various user requests and preferences for the bi -directional broadcast environment are increasing. Automatic video summary is needed for an efficient management and usage of many contents in service provider as well. In this paper, we propose a method to generate a content-based summary of linguistic educational videos automatically. First, shot-boundaries and keyframes are generated from linguistic educational video and then multiple(low-level) visual features are extracted. Next, the semantic parts (Explanation part, Dialog part, Text-based part) of the linguistic educational video are generated using extracted visual features. Lastly the XMI- document describing summary information is made based on HieraTchical Summary architecture oi MPEG-7 MDS (Multimedia I)escription Scheme). Experimental results show that our proposed algorithm provides reasonable performance for automatic summary of linguistic educational videos. We verified that the proposed method is useful ior video summary system to provide various services as well as management of educational contents.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.34 no.6 s.119
• /
• pp.54-65
• /
• 2007

There is little research on the exterior spaces or gardens of single detached urban dwellings in Korea because apartments are the general housing type of most inhabitants. However there is an increasing demand for idyllic houses in the urban fringe and for environmentally friendly dwellings or healthy houses that is represented in urban residents gardening activities. The purpose of this research is to analyze residents' preferences and problems in gardening and management of single detached urban dwellings, and to evaluate the possibility of gardens as biotopes. It also addresses the issue of providing public amenities in the urban landscape. One hundred thirty-six dwellings were selected in residential areas of the city of Gwangju. The layout and plant composition, residents' evaluation and understanding of their own gardens, and maintenance and management were investigated. The results indicated that residents associate gardening with mental health and recognized a lack of space as one environmental restriction. Watering was seen as a primary difficulty for maintaining gardens. This research suggests that providing space for gardening could be a tool for participatory community making. When considering the important aspects of gardening, residents responded that gardens offer naturalness, scenic beauty, and practicality and also felt that gardens provide a buffer from environmental pollution and relief from mental duress. Results from the factor analysis using 15 semantic differential variables showed that 'spatial openness', 'natural variety' and 'familiarity' were representative factors. Although a garden is mainly considered as a space for the appreciation of nature with plant material, our results suggest that meaning and environmental symbolism are important elements.

• Journal of the Korean Society for information Management
• /
• v.30 no.2
• /
• pp.227-243
• /
• 2013

The purpose of this study is to verify the correlation between the appearance frequency of author keyword and the number of citation in journal articles. In this study, we were trying to develop a methodology that can select the term having semantic relation with other terms and higher utilization to build a structured scientific glossary. In order to achieve this purpose, we analyzed the number of citation and the author keyword of the humanities and social science journal articles of the Korea Citation Index (KCI) from 2007 to 2011. This study found a correlation between appearance frequency of author keyword and the number of citation of the journal articles, with higher appearance frequency of author keyword of the journal articles being more cited.

• Journal of the Korean Society for information Management
• /
• v.33 no.3
• /
• pp.155-176
• /
• 2016

This study aims to develope a scholarly metadata information system based on conceptual elements of text structure of Korean studies research articles and to identify the applicability of text structure based metadata as compared with the existing similar system. For the study, we constructed a database(Korean Studies Metadata Database, KMD) with text structure based on metadata of Korean Studies journal articles selected from the Korea Citation Index(KCI). Then we verified differences between KCI system and KMD system through search results using same keywords. As a result, KMD system shows the search results which meet the users' intention of searching more efficiently in comparison with the KCI system. In other words, even if keyword combinations and conditional expressions of searching execution are same, KMD system can directly present the content of research purposes, research data, and spatial-temporal contexts of research et cetera as search results through the search procedure.

• Journal of the Korean Society for information Management
• /
• v.35 no.3
• /
• pp.165-187
• /
• 2018

Despite repeated efforts to develop a methodological foundation for assembling collaborative authority data in South Korea, issues such as the establishment of a standard authority model and standard authority construction as well as the reconfiguration of existing entities in authority building have prevented such research from generating a cooperative push for nation-wide authority data and progressing toward concrete implementation. The formulation of a collaborative and well-utilized collection of national authority data accordingly calls for 1) a practical approach to supporting both established authority data contributors and newly organized avenues of mutual participation in authority building, 2) committed involvement on the part of national institutions capable of providing the project with sustained assistance, and 3) a standard identification system which allows multiple organizations to merge their data. This study addresses the challenges of the current environment by taking stock of the key components necessary for the creation of collaborative authority data and using a Semantic Web-based interoperable VIVO ontology model to propose a viable national authority data framework.